AMD Ryzen 9 4950X "Vermeer" Tested, the Sample Boosts to 4.8 GHz

[A Computer Guy]

I own a 3950x watercooled and never it hits the 4.7. I would say more marketing BS.

I was curious to see how I could get my 3950x to hit 4.7 so I wrote some threads that added some integers one core at a time and it manged to hit 4.7 on core 0 , 4.5 on the rest of CCD1, and 4.3 all cores for CCD2.

 

[aQi]

Zen 3 with a combination of ddr5 would have been perfect. Since ddr5 is just around the corner. While ddr5 is already used in high-end mobile phones.

At least Threadripper should get ddr5 support. Supposingly an overkill but it sure adds interest to where amd is going as compared to Intel. (Not comparing AI)

 

[Makaveli]

Zen 3 with a combination of ddr5 would have been perfect. Since ddr5 is just around the corner. While ddr5 is already used in high-end mobile phones.

At least Threadripper should get ddr5 support. Supposingly an overkill but it sure adds interest to where amd is going as compared to Intel. (Not comparing AI)

Nah all new memory standards always come in slower than the previous one until given time to ramp up in speeds DDR5 will not be different.

It will also be more expensive.

Better to save it for the new Socket and Zen4 which is what I believe they are doing.

 

[net2007]

I was curious to see how I could get my 3950x to hit 4.7 so I wrote some threads that added some integers one core at a time and it manged to hit 4.7 on core 0 , 4.5 on the rest of CCD1, and 4.3 all cores for CCD2.

I tried to get 1 core 4.7 and the rest 4.4 on one CCD but the rest of the cores only went to 3.6 for some reason. However, with all cores set to 4ghz @ 1.3v It's a beast.

 

[ARF]

AMD Ryzen 9 4950X 16 Core 'Zen 3' Desktop CPU Now Spotted With Faster 4.9 GHz Boost Clocks In ES State, 200 MHz Faster Than The Ryzen 9 3950X

AMD Ryzen 9 4950X 16 Core Zen 3 CPU Spotted With Up To 4.9 GHz Clocks, 200 MHz Higher Clocked Than The Ryzen 9 3950X

The AMD Ryzen 9 4950X AM4 Desktop CPU which is powered by 16 Zen 3 cores has been spotted once again but with higher 4.9 GHz clock speeds.

wccftech.com

 

[Rob94hawk]

It really is! This point in time reminds me of 2004/2005/2006 time frame.

Built a rig just so it could play Doom 3!
Still have all the parts to rebuild! FX-53/X800XT. Good times.

 

[robot zombie]

At this point, I'm really not getting the obsession with boost clocks. I don't care what the damned thing clocks... it's still a new design. The question is how it performs at those clock speeds. I feel like by now everyone should know well and good that while increasing clocks on any particular chip will increase performance, it doesn't translate across different CPU's. One won't outperform the other because it clocks faster. Every single aspect of a CPU's functionality has compromises. Boost one, lose another. And I'm not talking about simple perf/temp tradeoffs, though that's one example.

AFAIC, boost clocks are just a number until they get attributed to measured performance.

 

[stimpy88]

Disappointing, but I guess we also have a +-15% IPC boost to go along with a tiny 100MHz bump.

It's a shame, as I was expecting AMD to be part of the 5GHz club, to shut the other team up, as we all know the vast majority think the faster the frequency, the better.

 

[nemesis.ie]

I presume you mean ~15%, -15% would not be good.

 

[Super XP]

What a joke. AMD literally are doing the same as Intel. Same CPU with little higher clocks and new model number.

ZEN3 is based on a New Micro-Architecture. aka New Design with a completely revamped cache system to reduce or eliminate latency hits.
If they can achieve this, my wallet will be

 

[ARF]

This news is a proof that the next is not and cannot be real:

AMD Next Generation ‘Vermeer’ Processors To Be Called Ryzen 5000 Series, May Max Out At 12 Cores Instead Of 16

AMD Next Generation 'Vermeer' Processors To Be Called Ryzen 5000 Series, May Max Out At 12 Cores Instead Of 16

According to a leak by an unknown source (via Videocardz), AMD will be utilizing 5000 series naming instead of 4000 series naming for its upcoming Zen-3 based Vermeer series of processors. This is something that makes a lot of sense considering AMD already has Ryzen 4000 series on mobility...

wccftech.com

 

[Super XP]

AMD should never have confused the naming schemes for desktop discrete Ryzen and integrated Ryzen with the GPU. It's confusing for the majority of people that are not up to date.
When you compare the Ryzen 3700X for example with a Ryzen 4000 mobile, people would assume that the Ryzen 4000 mobile is a newer design but it's not, both are based on ZEN2. One has the graphics integrated.

Not sure why AMD chooses to do this, because it ends up complicating its product line and messes up there naming scheme for future products.

To distinguish mobility and desktop discrete, AMD probably should have called these mobile CPUs Ryzen M3000 or something and ZEN3 mobile as Ryzen M4000.

 

[ARF]

AMD should never have confused the naming schemes for desktop discrete Ryzen and integrated Ryzen with the GPU. It's confusing for the majority of people that are not up to date.
When you compare the Ryzen 3700X for example with a Ryzen 4000 mobile, people would assume that the Ryzen 4000 mobile is a newer design but it's not, both are based on ZEN2. One has the graphics integrated.

Not sure why AMD chooses to do this, because it ends up complicating its product line and messes up there naming scheme for future products.

To distinguish mobility and desktop discrete, AMD probably should have called these mobile CPUs Ryzen M3000 or something and ZEN3 mobile as Ryzen M4000.

Technically, the mobile lineup with its monolithic designs is superior to the desktop chiplet offerings.
It makes sense that Zen 2 mobile monolithic is Ryzen 4 U, while Zen 2 desktop with worse power consumption chiplets is Ryzen 3 X.

The topic here is something else, though.

How on Earth would they allow themselves NOT to give any performance upgrade going from the 16-core Ryzen 9 3950X to a potential 12-core Ryzen 4000X, much worse if it end up Ryzen 5000X?
It'd be disastrous performance regression not progress or any improvement

 

[Super XP]

Technically, the mobile lineup with its monolithic designs is superior to the desktop chiplet offerings.
It makes sense that Zen 2 mobile monolithic is Ryzen 4 U, while Zen 2 desktop with worse power consumption chiplets is Ryzen 3 X.

The topic here is something else, though.

How on Earth would they allow themselves NOT to give any performance upgrade going from the 16-core Ryzen 9 3950X to a potential 12-core Ryzen 4000X, much worse if it end up Ryzen 5000X?
It'd be disastrous performance regression not progress or any improvement

The topic here is something else?
FYI, I was on topic.

And I wasn't speaking about performance, I was speaking about generational naming schemes. Mobile seems to keep getting the back seat on that, and it shouldn't. Comprenday?

 

[ARF]

The topic here is something else?
FYI, I was on topic.

And I wasn't speaking about performance, I was speaking about generational naming schemes. Mobile seems to keep getting the back seat on that, and it shouldn't. Comprenday?

Nope, mobile is called U, H, and HS, desktop is called X or non-X. Simple as this.

 

[Super XP]

Nope, mobile is called U, H, and HS, desktop is called X or non-X. Simple as this.

I'm going to upgrade my system with a new Ryzen 4800U because I want ZEN3.

 

[ARF]

I'm going to upgrade my system with a new Ryzen 4800U because I want ZEN3.

Nope, Zen 3 will be Ryzen 5000 U, H or HS and Ryzen 4000 X or non-X

 

[Super XP]

Nope, Zen 3 will be Ryzen 5000 U, H or HS and Ryzen 4000 X or non-X

You just proved my point I was talking about. Naming schemes are F'ed Up for those that don't understand them.

I had to stop a co worker from upgrading his Ryzen 1800X with. Ryzen 4800 because he's been waiting for ZEN3 for over a year. And didn't think upgrading to ZEN2 was sufficient to justify a upgrade.

If I wasn't there to stop him, when AMD actually releases ZEN3 this guy would have been pissed off. Understand? Lol

 

[ARF]

You just proved my point I was talking about. Naming schemes are F'ed Up for those that don't understand them.

I had to stop a co worker from upgrading his Ryzen 1800X with. Ryzen 4800 because he's been waiting for ZEN3 for over a year. And didn't think upgrading to ZEN2 was sufficient to justify a upgrade.

If I wasn't there to stop him, when AMD actually releases ZEN3 this guy would have been pissed off. Understand? Lol

No.
A person who understands what Zen 3 is, will for sure know what's under the naming scheme and numbers.

This difference comes from that originally Ryzen 1 launched in March 2017, while the mobile lineup was delivered much later, much closer to the physical Ryzen 2000 desktop availability.
It doesn't make sense to launch products with time gaps of half a year or more and call it in the same grouping.

 

[Super XP]

No.
A person who understands what Zen 3 is, will for sure know what's under the naming scheme and numbers.

This difference comes from that originally Ryzen 1 launched in March 2017, while the mobile lineup was delivered much later, much closer to the physical Ryzen 2000 desktop availability.
It doesn't make sense to launch products with time gaps of half a year or more and call it in the same grouping.

Once again you proved my point. 99% of the population knows Jack about CPUs and rely on naming schemes. Such as higher number should mean newer design and performance.

I've proved my point as factual. We will agree to disagree.

 

[Aquinus]

No.
A person who understands what Zen 3 is, will for sure know what's under the naming scheme and numbers.

Uh, the average person understands that 2 is bigger than 1. As @Super XP said:

99% of the population knows Jack about CPUs and rely on naming schemes. Such as higher number should mean newer design and performance.

 

[ARF]

Uh, the average person understands that 2 is bigger than 1.

2 is bigger than 1 for one main purpose - to sell the mobile parts because otherwise people will think that the desktop is superior and will buy the notebooks less.
Hopefully you understand.

 

[Aquinus]

2 is bigger than 1 for one main purpose - to sell the mobile parts because otherwise people will think that the desktop is superior and will buy the notebooks less.
Hopefully you understand.

I disagree, that makes absolutely no sense.

 

[ARF]

I disagree, that makes absolutely no sense.

This is AMD's sales and marketing tactics, and the explanation why it's being used..

 

[Super XP]

This is AMD's sales and marketing tactics, and the explanation why it's being used..

I agree that you make a valid point, but that's not the point I was trying to make.

I'm only asking that AMD do a better job with its naming schemes because they are confusing for people that don't understand about CPU tech and rely on such names to understand.

I disagree, that makes absolutely no sense.

Agreed.

And seems TPU messed up my original post. As double quoting what I wrote. Lol it's fixed now